FIG. 1 is an example of a configuration of a conventional computer system. The file output from the application program for a simulation executed by a large scale computer 103 is normally stored in a file server device 102. The large scale computer 103 can perform various applications for simulation such as scientific computation etc., and the large scale computer 103 is connected to the file server 102 on a high-speed network. The file server 102 is configured using, for example, a magnetic storage device.
On the other hand, a visualizing computer 101 is normally provided separate from the large scale computer 103, and executes a visualizing application program (hereinafter referred to as a visualizing application), thereby accessing the file server 102. A user transfers a file selected through a visualizing application to the visualizing computer 101, thereby acquiring preferable data.
The application for simulation on the large scale computer 103 divides a simulation result and outputs the resultant divisions to a plurality of data files by spatially dividing an event to be simulated. A time-base dividing process can also be performed on these data files.
The visualizing application on the visualizing computer 101 reads data files divided on a time base, on a spatial base, and for each physical amount, and displays on a screen the event described on a data file in various methods.
In the computer system above, when a huge amount of data of terabyte or petabyte order is output from the large scale computer 103, it is necessary to read the huge amount of data from the file server 102 and visualize the data. In this case, there occurs the problem of a long entire processing time from the data output by the large scale computer 103 to the visualization by the visualizing computer 101 due to the access speed of the magnetic storage device configuring the file server 102 and a necessary file selecting operation, thereby incurring a higher cost.
Therefore, a user of the large scale computer 103, especially a researcher and a computation engineer for scientific computation have a strong demand to reduce the cost.
When data to be displayed is stored in a display device, a conventional technique includes a data processing device for displaying data of a file by transmitting a selection code corresponding to the file to the display device. In addition, there is a simulation result display device for selecting data from a sequence of data of signal processing simulation results, reducing the number of pieces of data, and displaying waveforms.    Patent Document 1: Japanese Laid-open Patent Publication No. 02-100124    Patent Document 2: Japanese Laid-open Patent Publication No. 09-091316